148

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

EXPERIMENTAL STUDY OF LOCAL BEHAVIOR OFSTRENGTHENED REINFORCED CONCRETE SHORTCORBEL BY BONDING CARBON FIBER FABRICS

Ivelina Ivanova1* and Jules Assih1

1 Laboratory of Civil Engineering, University of Reims Champagne-Ardenne, France.

*Corresponding author:Ivelina Ivanova ivelina_ivanova@hotmail.com

ISSN 2319 – 6009 www.ijscer.comVol. 4, No. 1, February 2015

© 2015 IJSCER. All Rights Reserved

Int. J. Struct. & Civil Engg. Res. 2015

Research Paper

INTRODUCTIONMost of the structures in civil Engineering, after50 years old, meet the current safety standardsor have excessive cracks. Steel corrosion mayalso cause the occurrence of high deflectionor instability of the structure itself. It is generallymanifested by poor performance under serviceloading in the form of excessive deflections orcracking.

The introduction about 34 years ago ofcomposite materials in the field of CivilEngineering allows other strengthening or

This paper presents the local behavior study of strengthened reinforced concrete short corbelby bonding composite carbon fiber fabrics. Indeed, reinforced concrete short corbels often areused in civil engineering, in buildings. The effect of some parameters on mechanical behavior ofstructure strengthening is examined. As such as, influence of the type of strengthening by gluingdirectly on concrete corbel or wrapping. The used fiber fabrics are unidirectional and bidirectionalcarbon fabrics. The ultimate load and displacement of specimen are compared to the referencestructure of static tests monotonous applied load. The cracking and failure mode under staticloading are presented. The extensometer technique based on deformation gauge strain is usedto study the local behavior of the structure. This technique allows to measure deformations ofgauges glued to the surface of the composite fabrics of carbon fibers bonded, of the concreteand of steel.

Keywords: Strengthening, Bonded, Damaged, Carbon, Fabrics, Behavior

repair of reinforced concrete structures bybonding composite carbon fiber fabrics (AbdulWahab, 1989; ACI, 2000; Chris, 2007).Carbon fiber materials have many advantages(Ivanova, 2013): their weight, flexibility,implementation easier and also theirphysicochemical properties (corrosion)interesting. This technical of strengtheningcompensate the loss of rigidity and resistanceto cracking due to the strengthening andimproving performance and durability ofstructures.

Corbel is one important element of structure

149

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

to support the pre-cast structural system suchas pre-cast beam and pre-stressed beam(Anis, 2012 and Rejane, 2005). The corbel iscast monolithic with the column element or wallelement. It is interesting to study the localmechanical behavior of this very short elementof the structure using carbon fiber materials(Mohammed, 2005; Futtuhi, 1987; Gampione,2005; Erfan, 2010).

This paper is mainly interested in the studyof three types of reinforcement: by bondedcarbon fiber fabrics, wrapping of carbonfabrics and by bonding plate in shear area,under flexural bending. Local deformationusing strain gauges to measure strains in thesteel, concrete and carbon fiber sheets ofstrengthened reinforced concrete short-corbel,is also investigated. In this investigation,deformations, cracking modes and collapsemechanism are studied.

EXPERIMENTAL PROGRAMThis technic for carrying out such improvementwas that which involved bonding of steel platesto structure surfaces. An effective way ofeliminating the corrosion problem was toreplace steel plates with corrosion resistancematerials such as fiber composite materials.

Many advantages are: low density, corrosion,mechanical properties, good resistance tofatigue and ease of handling.

MaterialsNormal strength concrete materials are rolledgravel dried sand and ordinary Portlandcement. The cement:sand:gravel proportionsin the concrete mix were 1:1.73:2.93 by weightand the water/cement ratio was 0.50. Portlandcement type CEM II was used and themaximum size of the aggregate was 12.5 mm.

Four 200 x 200 x 200 mm3 concrete cubicwere also cast and are tested when each shortcorbel is tested to determine the compressivestrength of the concrete at 28 days of age.

The glue used for the CFC sheet bondingtechnique are generally two part systems, aresin and a hardener, and when mixed. Theelastic modulus and yield stress are presentedin Table 1.

Steel bars, S500 are used of differentdiameters: 6, 10, 14 mm. The steel specimensare characterized by simple testing tensile.The stress fu and the modulus of elasticity Es

values are in Table 1. The high deformation ofthis steel at the failure is 11.04%.

Table 1: Properties of Materials

Material Young’s Modulus (GPa) Strength (MPa) Poisson Ratio

Concrete 30±2 33,2±2 (fc) 0,25

Steel bar 200±1 610±10 (fu) 0,30

Adhesive 4,1±1 36±1 (fu) 0,41

UCFC sheet* 86±1 1035±63 (fu) 0,45

BCFC sheet** 87±1 720±50 0,35

Note: * UCFC sheet: Unidirectional Carbon fiber composite sheet; ** BCFC sheet: Bidirectional Carbon fiber Composite sheet.

150

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

The unidirectional and bidirectional fibersare used. The experimental results obtainedfor carbon composite unidirectional areshowed in Table 1.

The yield strength corresponds to the tensilestrength. The carbon composite sheets havea linear elastic behavior up to failure. The highelongation at failure is 0.8% for theunidirectional carbon fiber sheet and 0.5% forthe bidirectional carbon fiber sheet. Table 1shows the characteristic properties of usedcomposite materials.

The surface preparation was of primaryimportance and calls for care (Figure 1).Preparation of concrete surfaces must becarried out to remove any loose or weakmaterial, oil, grease, etc. In this case gritblasting was being the good method. The fourcorners of the corbel are rounded to reducethe decrease in strength and to prevent tearingof the composite material. Preparation ofsurface should be carried out just prior to thebonding operation to prevent any

and carbon fiber fabrics is applied with thebrush.

The concrete surface has already becomeroughened and then leveled before sticking onthe wraps using epoxy adhesive. Pressuremust be applied to squeeze out excess glueand held the plate in place until the glue hashardened. Resin and hardened glue is justmixed before the gluing operation.

Experimental DeviceAll the corbels are tested under tree-pointsload. The specimen is submitted to a verticalload equivalent to the response of the bearingwhich is half of the load jack. All tests areperformed with a loading speed average of0.2 kN/s. The registration system “systemVishay” data is recorded every 0.1 s. The loadcapacity of the test bending is 1000 kN.

Electric GaugesWe present the result of extensometertechnique based on deformation gauges,strain versus applied load used to study thelocal behavior of structure. Strains weremeasured by strain gauges of 120 andFigure 1: Surface Treatment

by Sandblasting

contamination. After, the contamination can beavoided by applying the glue to the concrete

Figure 2: Position of electric gauges onunderwood of reinforced concrete corbel

151

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

gauge factor 2.09. The strain gauges areattached to each of the experimentalspecimen. Precision strain gauges length of5 mm (Figure 3) are used to measure the strainon the main steel reinforcement (Figure 2) aswell as horizontal frame, steel tie rod placedalong the tension and the compression strut.The strain gauges on the face of the corbel areplaced in shear span between the corbel andthe column. Two of gauges length of 10 mm areplaced on the composite fibers plate surfaceand one gauge length of 30 mm on the concrete.

Interpretation of curvesThe curves F=f() describing the local behaviorof the composite plate, steel and concretematerial is presented (F is the applied loadand the strain indicated by the gauge locatedat x). The change of slope dF/d correspondsto the appearance of micro cracks or a changeof state. The change of sign of dF/dcorresponds to the initiation of the rapid crackpropagation around of the ultimate failure ofthe structure.

The control specimen without strengtheningis denoted “C0”, the letter “C” means Corbel,zero 0 indicate without strengthening.

For the designation of the strengthenedreinforced concrete corbel, the first letter “C”is, as previously, Corbel, and the second letterrepresents the type of strengthening (e.g.: Pfor Plate, B for Bandage, wrapping). Then, digitindicates the number of layers (e.g.: 1, 2, 3,5). Finally the small letter indicates the type ofcomposite material (e.g. u for Unidirectionaland b for Bidirectional).

RESULTS AND DISCUSSIONFrom the large amount of data obtained fromtesting, only typical data are reported here inTable 2. This paper present the results on thetype of strengthening on bending and shear,bonding directly carbon fiber fabrics on eithersurfaces or full wrapping of short reinforcedconcrete corbel.

Four concrete corbels were tested. Threeof them are strengthened bonding carbon fiberfabric as in Table 2 and one control specimenwithout reinforcing.

The results show that there is a verysignificant increase in capacity loading for thethree types of reinforcement 41 to 82%.

Figure 3: Detail of bonded and welding of electric gauge

152

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

FR: Ultimate load of strengthenedreinforcement concrete corbel.

F0: Ultimate load of unstrengthenedreinforcement concrete corbel.

The ultimate strength of reference corbel isF0, equal to 357 kN. The results show anincrease in the load FR/F0 to 1.41 for CI2b to1.82% for CB3u. In the former case, the plates45 are bonded to the sewing degrees obliquefissure. The bonded fabric of carbon fibers arearepresents 32.5% of corbel total area. Whilethe second case is that of strengthening then,represents the area 51, 6%. In the latter case,the complete tire of the console, this area is60.0%. The results show that strengthening ismore effective wrapping trough the confiningeffect of concrete and a resumption of effortsby carbon fiber fabric. Doubling the bondingsurface, the tensile strength is increased by

14%. Of course, in the case of the containment,are double ultimate load. Accordingly, the testresult shows, that composite plates judiciouslysuitable on surface more effectively, loadcapacity is increased.

Figure 4 shows the deformation in recessedsection of the corbel. Deformations are higherin the tension of the console until mid-height,so bond a plate that you can reconnect effortsmore effectively.

The results shown in Figure 5 thecomparison of local steel deformation in crosssection of strengthened reinforced concretecorbels CB3u, CP3u, CI2b and C0.Containment effect is evidenced by thereduction of deformations and a third of chargemore than twice between Cb3u and C0. In thecase of strengthening, CI2b configurationappears effective, insofar as crack is sewn by

Table 2: Design of Testing Specimen

FR/F0 Design Designation

1,41% Strengthened corbel by bonded Inclined CI2bbidirectional carbon fiber fabrics on both sides

1,55% Strengthened corbel by bonded unidirectional CP3ucarbon fiber fabrics on both sides.

1,82% Wrapping strengthened corbel by bonded CB3ucomposite plates on both sides.

Note: FR: ultimate load of strengthened reinforcement concrete corbel; F0: ultimate load of unstrengthened reinforcement concrete corbel

153

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

bonded carbon fiber fabrics to 45°.Confinement effect disappears as in the other

two cases. Strains are more important (~7000µm/m). Recovery strengths in the tie rod steel

Figure 4: Local Strain of Steel in Cross Section of Reinforced Concrete Corbel

Figure 5: Comparison of Steel Strain in Cross Section (point G1) of Strengthened Corbel

154

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

causes a local yielding which causes highdeformations. Strengthened reinforcedconcrete corbel by bonded carbon fiber fabricstrips to 45° has a ductile mechanical behavior.However, the result that CP3u shows a similarbehavior as CB3u, provided to ensure ananchor to prevent detachment of the plate.

Specimen CP3uThis specimen was strengthened withexternally bonded tree horizontal fabric strip.There is only one major diagonal crack startedat the bearing plate, and propagated towardsthe junction of the column and face of thecorbel. This crack appears at a load of 310kN and causes failure of the corbel. The corbelfailed at an ultimate load of 380 kN. When thenumber of fabric strips is increased, theultimate deformation of the corbel is causedby peeling off the strips of cloth. Theaccumulation of several strips of fabric on topof each other results in a thickening of the layeras a whole, and therefore, they peel off easilyand cannot reach to the maximum load-bearingcapacity of the composite. There was noapparent damage to the fabric, just concretecracked major diagonal shear crack beneaththe fabric. The concrete crack caused ofdebonded fabrics on the front and rear face ofthe corbel.

Specimen CB3uFirst flexural cracks appear at 140 kN and thenappearance of the bearing at 365 kN. Probablythe appearance of diagonal cracks plays a lotbecause of the positive move in after thelanding. It shows the comparison of curvesobtained by gauges “G1” (steel tie localdeformation) for four different corbels. Thesecurves are compared to reference reinforced

concrete corbel without strengthening. Theresults show that curves are similar andultimate load increases with a third reductionof deformation. The failure and crack patternsof the group strengthening with externallybonded one horizontal strip of fabric. In all fourcases, first micro cracks is started atapproximately the same load between 130and 140 kN. There is one main diagonal shearcrack almost at an angel 45 degrees and thiscrack is started at the bearing plate. This crackcaused failure of the corbel. Corbel failed atultimate loads at 651 kN.

Specimen CI2bFirst flexural cracks appear at 116 kN. As withthe above specimen CP3u, there is only onemajor diagonal crack started at the bearingplate, and propagated towards the junction ofthe column and face of the corbel. This crackappears at a load of 380 kN and causes failureof the corbel. The corbel failed at an ultimateload of 520 kN. There was no apparentdamage to the fabric, just concrete crackedmajor diagonal shear crack. The collapse ofCI2b is appeared by Shear failure diagonalcrack of CFRP fabric.

Position of the Neutral AxisConcrete surface strains at a section close tocolumn face of the tested corbels weremeasured by using three strain gauges over thedepth of the corbel. Figures.6 and 7 show theconcrete strain distribution over the corbel depthfor all corbels at different load levels. In thesefigures, it can be seen that the strain distributionwas approximately linear in tensile andcompressive zones at low load levels, and thenbecame increasingly nonlinear at the tensionzone at higher loads due to cracking effect.

155

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

For all tested corbels, the neutral axis wasshifting upwards after increasing the loadbeyond than the load at first crack. The positionof the neutral axis for small load (10 kN) isabout 15 cm, for which reason the width of theplate is less than or equal to this value.

Modes of FailureThe results, as shows in Figure 8(a), 8(b), 8(c)and 8(d), distinguished mainly three types offailure: flexion failure, shear and fabrics peelingoff and compression failure. Although, CI2b andC0 were the same rupture, noted that cracking

Figure 7: Position of Neutral Axial Versus Applied Load for CB3u

Figure 6: Position of Neutral Axial Versus Applied Load for CP3u

156

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

Figure 8: Different Failure Modes of Corbels

a) C0 : Flexion failure b) CB3u : Compression failure

c) CP3u : Shear collapse with fabrics peeling off d) CI2b : Flexion and little compression failure

157

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

was different with appearance of a slightcompression in CI2b.

CONCLUSIONThe contribution of strengthening reinforcedconcrete short corbels is very significant andinteresting. Result shows an increase in failuretensile strength more than 1.82 by bondingcarbon fiber fabric.

The bonding composite carbon fiber fabricsslightly changes the position of neutral axis.But for loads less than 150 kN, the straindistribution in cross section appears linear.Beyond this value, the strain distribution is non-linear as in Figure 6 and 7.

Result show that strengthening by wrappingremains still best configuration. Effect ofcontainment by wrapping provides highervalues than those at which the plates are glueddirectly on the front breaking loads.

Corbel CI2b the configuration presents aductile behavior and avoids the sudden shearfailure of unstrengthened reinforced concretecorbel.

There are mainly three types ofstrengthened reinforced concrete corbelfailures: flexion failure, shear and fabricspeeling off failure, compression failure with twodiagonal cracks, figure 8.

This article shows that it is possible toenhance or repair a reinforced concretestructure by bonding composite carbon fiberplate. This technique provides better stressdistribution and implementation easier even ifit requires careful preparation.

REFERENCES1. Abdul-Wahab H M (1989), “Strength of

Reinforced Concrete Corbels With

Fibers”, ACI Structural Journal, Vol. 86,No. 1, pp. 60-66.

2. ACI Committee 440 (2000), “Guide forthe Design and Construction of ExternallyBonded FRP Systems for StrengtheningConcrete Structures,” American ConcreteInstitute, Farmington Hills, Mich., January,p. 79.

3. Chris Burgoyne and Ioannis Balafas(2007), “Why is FRP not a financialsuccess, FRPRCS-8 University ofPatras”, Greece, July 16-18.

4. Ivanova I. (2013), Mechanical behaviorof short concrete corbels reinforced orrepaired by bonding of compositematerials, Ph.D. thesis, URCA, France.

5. Anis A Mohamad-Ali and MuhammadAbed Attiya (2012), “Experimentalbehavior of reinforced concrete corbelsstrengthened with carbon fibre reinforcedpolymer strips”, Basrah Journal forEngineering Science.

6. Rejane Martins Fernandes and MounirKhalil El Debs (2005), “Análise dacapacidade resistente de consoles deconcreto aramado considerando acontribuição da armadura de costura.Cadernos de Engenharia de Estruturas”,São Carlos, Vol. 7, No. 25, pp. 103-128.

7. Mohamed A Elgwady, Mohamed RABIÉand Mohamed T Mostafa (2005),“Strengthening of corbels using CFRP anexperimental program, Building andStructural engineering”, Cairo University,Giza, Egypt.

8. Futtuhi I Nijad (1987), “SFRC Corbel

158

Int. J. Struct. & Civil Engg. Res. 2015 Ivelina Ivanova and Jules Assih, 2015

Tests”, ACI Structural Journal, Vol. 84,No. 2, pp. 11981213.

9. Gampione G, Mendola L L and Papia M(2005), “Flexural Behaviour of ConcreteCorbels Containing Steel Fibers orWrapped with FRP Sheets,” Materials

and Structures, Vol. 38, pp. 617-625.

10. Erfan A M (2010), “Behavior ofReinforced Concrete CorbelsStrengthened with CFRP Fabrics,” M.Sc.Thesis, Department of Civil Engineering,Benha University, Shoubra, Egypt.